gh-142236: Improve error location for missing comma in string concatenations (#142330)

diff --git a/Grammar/python.gram b/Grammar/python.gram
index 7ae00c6f005e7e3681e005e9fc2bb98c57e63bab..110136af81b59638b989b31870d91063082bbe66 100644 (file)
--- a/Grammar/python.gram
+++ b/Grammar/python.gram
@@ -1251,8 +1251,7 @@ invalid_expression:
     # !(NAME STRING) is not matched so we don't show this error with some invalid string prefixes like: kf"dsfsdf"
     # Soft keywords need to also be ignored because they can be parsed as NAME NAME
    | !(NAME STRING | SOFT_KEYWORD) a=disjunction b=expression_without_invalid {
-        _PyPegen_check_legacy_stmt(p, a) ? NULL : p->tokens[p->mark-1]->level == 0 ? NULL :
-        RAISE_SYNTAX_ERROR_KNOWN_RANGE(a, b, "invalid syntax. Perhaps you forgot a comma?") }
+        _PyPegen_raise_error_for_missing_comma(p, a, b) }
    | a=disjunction 'if' b=disjunction !('else'|':') { RAISE_SYNTAX_ERROR_KNOWN_RANGE(a, b, "expected 'else' after 'if' expression") }
    | a=disjunction 'if' b=disjunction 'else' !expression {
         RAISE_SYNTAX_ERROR_ON_NEXT_TOKEN("expected expression after 'else', but statement is given") }

diff --git a/Lib/test/test_syntax.py b/Lib/test/test_syntax.py
index ade8f273a1e08871293c4e80423066a633d61230..93f0b98de71d8102f8011621485ebfa54ee82de9 100644 (file)
--- a/Lib/test/test_syntax.py
+++ b/Lib/test/test_syntax.py
@@ -3336,6 +3336,20 @@ case(34)
             lineno=3
         )
 
+    def test_multiline_string_concat_missing_comma_points_to_last_string(self):
+        # gh-142236: For multi-line string concatenations with a missing comma,
+        # the error should point to the last string, not the first.
+        self._check_error(
+            "print(\n"
+            '    "line1"\n'
+            '    "line2"\n'
+            '    "line3"\n'
+            "    x=1\n"
+            ")",
+            "Perhaps you forgot a comma",
+            lineno=4,  # Points to "line3", the last string
+        )
+
     @support.cpython_only
     def test_syntax_error_on_deeply_nested_blocks(self):
         # This raises a SyntaxError, it used to raise a SystemError. Context

diff --git a/Misc/NEWS.d/next/Core_and_Builtins/2025-12-06-00-38-37.gh-issue-142236.m3EF9E.rst b/Misc/NEWS.d/next/Core_and_Builtins/2025-12-06-00-38-37.gh-issue-142236.m3EF9E.rst
new file mode 100644 (file)
index 0000000..a8d37b4
--- /dev/null
+++ b/Misc/NEWS.d/next/Core_and_Builtins/2025-12-06-00-38-37.gh-issue-142236.m3EF9E.rst
@@ -0,0 +1,3 @@
+Improve the "Perhaps you forgot a comma?" syntax error for multi-line string
+concatenations to point to the last string instead of the first, making it
+easier to locate where the comma is missing. Patch by Pablo Galindo.

diff --git a/Parser/action_helpers.c b/Parser/action_helpers.c
index b7a5b9d5e307b1d085607c904a8440f3e2aa6633..50856686335a147b072b376253db9e6749d4ea15 100644 (file)
--- a/Parser/action_helpers.c
+++ b/Parser/action_helpers.c
@@ -947,6 +947,35 @@ _PyPegen_check_legacy_stmt(Parser *p, expr_ty name) {
     return 0;
 }
 
+void *
+_PyPegen_raise_error_for_missing_comma(Parser *p, expr_ty a, expr_ty b)
+{
+    // Don't raise for legacy statements like "print x" or "exec x"
+    if (_PyPegen_check_legacy_stmt(p, a)) {
+        return NULL;
+    }
+    // Only raise inside parentheses/brackets (level > 0)
+    if (p->tokens[p->mark - 1]->level == 0) {
+        return NULL;
+    }
+    // For multi-line expressions (like string concatenations), point to the
+    // last line instead of the first for a more helpful error message.
+    // Use a->col_offset as the starting column since all strings in the
+    // concatenation typically share the same indentation.
+    if (a->end_lineno > a->lineno) {
+        return RAISE_ERROR_KNOWN_LOCATION(
+            p, PyExc_SyntaxError, a->end_lineno, a->col_offset,
+            a->end_lineno, a->end_col_offset,
+            "invalid syntax. Perhaps you forgot a comma?"
+        );
+    }
+    return RAISE_ERROR_KNOWN_LOCATION(
+        p, PyExc_SyntaxError, a->lineno, a->col_offset,
+        b->end_lineno, b->end_col_offset,
+        "invalid syntax. Perhaps you forgot a comma?"
+    );
+}
+
 static ResultTokenWithMetadata *
 result_token_with_metadata(Parser *p, void *result, PyObject *metadata)
 {

diff --git a/Parser/parser.c b/Parser/parser.c
index 648b3702d8ff71b1b3d6a741b1d867a4ece13638..09bfb5725a2ec31a79b2aa30083b5c3b0a54306a 100644 (file)
--- a/Parser/parser.c
+++ b/Parser/parser.c
@@ -21445,7 +21445,7 @@ invalid_expression_rule(Parser *p)
         )
         {
             D(fprintf(stderr, "%*c+ invalid_expression[%d-%d]: %s succeeded!\n", p->level, ' ', _mark, p->mark, "!(NAME STRING | SOFT_KEYWORD) disjunction expression_without_invalid"));
-            _res = _PyPegen_check_legacy_stmt ( p , a ) ? NULL : p -> tokens [p -> mark - 1] -> level == 0 ? NULL : RAISE_SYNTAX_ERROR_KNOWN_RANGE ( a , b , "invalid syntax. Perhaps you forgot a comma?" );
+            _res = _PyPegen_raise_error_for_missing_comma ( p , a , b );
             if (_res == NULL && PyErr_Occurred()) {
                 p->error_indicator = 1;
                 p->level--;

diff --git a/Parser/pegen.h b/Parser/pegen.h
index b8f887608b104eccbc96246ecd32e40a6d1e1f76..be5333eb2684ae2d3aec630ca175da34204ac0a1 100644 (file)
--- a/Parser/pegen.h
+++ b/Parser/pegen.h
@@ -358,6 +358,7 @@ expr_ty _PyPegen_ensure_real(Parser *p, expr_ty);
 asdl_seq *_PyPegen_join_sequences(Parser *, asdl_seq *, asdl_seq *);
 int _PyPegen_check_barry_as_flufl(Parser *, Token *);
 int _PyPegen_check_legacy_stmt(Parser *p, expr_ty t);
+void *_PyPegen_raise_error_for_missing_comma(Parser *p, expr_ty a, expr_ty b);
 ResultTokenWithMetadata *_PyPegen_check_fstring_conversion(Parser *p, Token *, expr_ty t);
 ResultTokenWithMetadata *_PyPegen_setup_full_format_spec(Parser *, Token *, asdl_expr_seq *, int, int,
                                                          int, int, PyArena *);